Determination of Nateglinide in Human Plasma by LC-ESI-MS and Its Application to Bioequivalence Study

Abstract

To evaluate the bioequivalence of nateglinide, a rapid and specific liquid chromatographic-electrospray ionization mass spectrometric
method was developed and validated to determine nateglinide for human plasma samples. The analyte was detected using electrospray
positive ionization mass spectrometry in the selected ion monitoring mode. Tinidazole was used as the internal standard. A
good linear relationship obtained in the concentration ranged from 0.05 to 16μgmL−1 (r
2=0.9993). Lower limit of quantification was 0.05μgmL−1 using 100μL of plasma sample. Intra- and inter-day relative standard deviations were 2.1–7.5 and 4.7–8.9%, respectively.
Among the pharmacokinetic data obtained, T
max was 2.09±1.06h for reference formulation and 2.40±0.97h for test formulation. C
max was 4.17±1.31μgmL−1 for reference formulation and 4.37±1.53μgmL−1 for test formulation. The half-life (t
½) was 1.93±0.44h for reference formulation and 1.92±0.29h for test formulation. AUC0–10h was 13.67±4.36μghmL−1 for reference formulation and 13.21±4.09μghmL−1 for test formulation. This method was successfully applied to the pharmacokinetic study in human plasma samples.

KeywordsColumn liquid chromatography-Mass spectrometry-Pharmacokinetic study-Nateglinide

Full text

Page 1

Determination of Nateglinide in Human
Plasma by LC-ESI-MS and Its Application
to Bioequivalence Study
De-En Han1, Yi Zheng1, Ning Li1, Di Zhao1, Ge Zhang1, Hefeng Yan2, Lingli Zhang1, Wei Sun1,
Ya-Ning Wu1, Yang Lu1, Xijing Chen1,&
1Center of Drug Metabolism and Pharmacokinetics, China Pharmaceutical University, Nanjing, China; E-Mail: chenxj@jlonline.com
2People’s Hospital of Lushan, Pingdingshan, China
Received: 6 December 2008 / Revised: 11 September 2009/ Accepted: 8 October 2009
Online publication: 22 December 2009
Abstract
To evaluate the bioequivalence of nateglinide, a rapid and specific liquid chromatographic-
electrospray ionization mass spectrometric method was developed and validated to deter-
mine nateglinide for human plasma samples. The analyte was detected using electrospray
positive ionization mass spectrometry in the selected ion monitoring mode. Tinidazole was
used as the internal standard. A good linear relationship obtained in the concentration
ranged from 0.05 to 16 lg mL-1(r2= 0.9993). Lower limit of quantification was
0.05 lg mL-1using 100 lL of plasma sample. Intra- and inter-day relative standard
deviations were 2.1–7.5 and 4.7–8.9%, respectively. Among the pharmacokinetic data
obtained, Tmaxwas 2.09 ± 1.06 h for reference formulation and 2.40 ± 0.97 h for test
formulation. Cmax was 4.17 ± 1.31 lg mL-1for reference formulation and 4.37 ±
1.53 lg mL-1for test formulation. The half-life (t½) was 1.93 ± 0.44 h for reference for-
mulation and 1.92 ± 0.29 h for test formulation. AUC0–10hwas 13.67 ± 4.36 lg h mL-1
for reference formulation and 13.21 ± 4.09 lg h mL-1for test formulation. This method
was successfully applied to the pharmacokinetic study in human plasma samples.
Keywords
Column liquid chromatography
Mass spectrometry
Pharmacokinetic study
Nateglinide
Introduction
Type II diabetes mellitus is a long-term
metabolic disorder and resistant to the
effects of insulin, a hormone that regu-
lates sugar absorption [1]. Antidiabetic
agents, such as sulfonylurea and thiazo-
lidinedione derivatives, are used for the
treatment of non-insulin-dependent type
II diabetes mellitus. Nateglinide [NTG,
N-(trans-4-isopropylcyclohexylcarbonyl)-
D-phenylalanine] is a
derivative designed to restore insulin
secretion in Type II diabetic patients [2].
NTG increases insulin release from
pancreatic b-cells through inhibition of
potassium-ATP channels [3]. After oral
administration, NTG is rapidly absorbed
and peak plasma concentrations are
reached in 0.5–1.0 h post dose [4, 5].
While evaluating
methods of nateglinide, we find several
methods for detection and quantitation
of anti-diabetic drugs in biological sam-
ples. A micellar electrokinetic chro-
matographic (MEKC)
on-line sweeping technique has been
developed for NTG determination in
D-phenylalanine
theanalytical
methodwith
2010, 71, 299–304
DOI: 10.1365/s10337-009-1405-4
0009-5893/10/02
? 2009 Vieweg+Teubner | GWV Fachverlage GmbH
Full Short CommunicationChromatographia 2010, 71, February (No. 3/4)
299

Page 2

animal plasma by Yan et al. [6]. Avail-
able reports suggest that liquid chroma-
tography–tandemmass
(LC–MS–MS) and LC with UV detec-
tion in conjunction with solid phase
extraction(SPE)
extraction (LLE) are the most commonly
used methods for detection and quanti-
tation of anti-diabetic drugs in biological
samples [4, 7–17]. NTG and its metabo-
lites along with other anti-diabetic drugs
has been identified in plasma by liquid
chromatography–tandem mass spectro-
metric detection using an ion trap mass
spectrometer equipped with electrospray
ionization interface or in the atmo-
spheric pressurechemical
mode [18–22]. The bioavailability of
NTG-hydroxypropyl-b-cyclodextrin com-
plex capsule in rabbits has been studied
by liquid chromatography–tandem mass
spectrometry [23].
However, they either have a limited
sensitivity or are time-consuming or re-
quire special equipments. In the present
article, we report a simple, precise and
sensitive liquid chromatographic-elec-
trospray ionization-mass spectrometry
(LC-ESI-MS) method for the quantita-
tive determination of NTG in human
plasma after protein precipitation. Sam-
ple preparation omits cumbersome SPE
or LLE methodologies, is automated,
and drug elution is achieved without
need of derivatization. Accurate quanti-
tation of NTG is achieved over a wide
range of concentrations. The method
was successfully applied to investigate
the bioavailability of NTG.
spectrometry
orliquid–liquid
ionization
Experimental
Chemicals and Reagents
NTG and tinidazole (IS) were obtained
from the National Institute for the
Control of Pharmaceutical and Biologi-
cal Products (Beijing, China). NTG
tablets (30, 120 mg per tablet) were from
Wan Bang Medicine Company (Jiangsu
Province, China).
grade) used in the mobile phase was
purchased from Tedia (Fairfield, USA).
Formic acid and other chemicals were of
analytical grade (Nanjing
Reagent CO., Jiangsu Province, China).
Acetonitrile(LC
Chemical
Liquid nitrogen was from the Gas
Supplier Center of Nanjing University
(Jiangsu Province, China).
Instrumentation
A Shimadzu (Kyoto, Japan) LC–MS-
2010A liquid chromatograph-mass spec-
trometer equipped with a SIL-HTC
autosampler,twoLC-10AVPpumps,and
anelectrospray-ionization(ESI)interface
wasusedforLC–MSanalyses.Analytical
data acquisition and processing were
accomplished using Shimadzu LC–MS
solution (Version 3.20 with Windows XP
operating system) Software.
LC–MS Conditions
LC separation was performed using
Shim-packstainless-steel
(HypersilODS2,
4.6 mm, Dalian Elite Company Liaon-
ing Province, China) coupled with a
SecurityGuard C18 guard column (4 9
3.0 mm, Phenomenex, Torrance, CA,
USA). The mobile phase consisted of
acetonitrile–water–formic acid (61:39:1,
v/v/v) and was delivered at a flow-rate of
1.0 mL min-1. The column temperature
was maintained at 40 ?C. A quadrupole
mass spectrometer equipped with an
electrospray ionization
operated with a gas (N2) flow of
1.5 L min-1and detector voltage of
1.80 kV. The heat block temperature
was 200 ?C and the curved desolvation
line (CDL) temperature was maintained
at 250 ?C. LC-ESI-MS was performed in
positive selected-ion monitoring mode.
NTG was detected at m/z 318.25 and
tinidazole at m/z 247.90. The quantifi-
cation was performed using peak areas.
column
150 mm 9
5 lM,
source was
Preparation of Standard and
Quality Control Solutions
Stock standard solutions of NTG were
prepared by weight at 10 mg in MeOH.
Standard andquality
samples were made from separate stock
solutions. Standard samples were pre-
pared over a range of 0.05–16 lg mL-1
of NTG. A working IS solution was
control(QC)
prepared at 25 lg mL-1. All standards
were stored at 4 ?C.
Sample Preparation
Human blood samples were centrifuged
at 4,0009g for 5 min immediately after
collection and 1 mL of the plasma was
promptly removed to an Eppendorf
tube. These samples were immediately
stored at -70 ?C until analysis. Aceto-
nitrile (400 lL) was added to protein
precipitation followed by the addition of
100 lL of human plasma and 10 lL of
tinidazole solution (IS). Then the tubes
were vortex-mixed and centrifuged at
10,000 rpmfor 10 min.
(0.2 mL) of the supernatant was then
removed, mixed with 0.05 mL of deion-
ized water and a 20 lL aliquot was
injected for analysis by LC-ESI-MS.
Analiquot
Extraction Recovery and
Matrix Effect
Extraction efficiency was expressed in
terms of recovered concentration of
analyte and IS added to a biological
matrix prior to extraction (recovery QC)
versus concentration obtained with a
biological sample where analyte and IS
were added following extraction (refer-
ence QC). All analyses were performed
in triplicate at three analyte concentra-
tions. Percent drug recovery with corre-
sponding %CV was determined for each
plasma sample fortified with the analyte.
To evaluate matrix effect, blank plasma
was subjected to sample pretreatment
described above. The resulting solution
was spiked with working standard solu-
tions to prepare solutions containing
NTG at three different concentrations
(0.1, 1.0 and8.0 lg mL-1).
enhancement/suppression of ionization
was evaluated by comparing the peak
areas of processed spiked samples with
corresponding neat standard solutions
prepared in mobile phase.
Matrix
Precision and Accuracy
The precision of the method was con-
firmedbythefivereplicatedeterminations
300
Chromatographia 2010, 71, February (No. 3/4)Full Short Communication

Page 3

of plasma samples containing NTG at
three concentration levels (0.1, 1.0 and
8.0 lg mL-1) in three different batches.
The determined concentrations, which
were obtained from a calibration curve
prepared on the same day, were used to
evaluate the method accuracy. The pre-
cision was evaluated by the coefficient of
variation (CV), and the accuracy was
expressedasthe
(RE) according to the equation: RE
(%) =100% 9 (measuredconcentration
- nominal concentration)/nominal con-
centration. The criteria for acceptability
ofthedataincludedprecisionwithin15%
CV and accuracy within 15% RE from
the nominal values.
relativeerror
Stability Experiments
Three concentration levels (0.1, 1.0 and
8.0 lg mL-1) of spiked samples were
assessed after the storage at room tem-
perature for up to 4 h, after three freeze–
thaw cycles, and at -20 ?C for one
month. NTG stability in the injection
solvent was determined periodically by
injecting processed samples for up to
12 h (in autosampler at 5 ?C) after the
initial injection. Samples were consid-
ered to be stable if the bias between the
concentrations determined from 0 h and
under different stability conditions was
within the acceptance criterion recom-
mended by the Food and Drug Admin-
istration (USFDA) (http://www.fda.gov/
cder/guidance/index.htm, 2001). The
bias was calculated as follows: % bias =
100 ? (C - C0 h)/C0 h
Application of the Assay to
Bioequivalence Study in
Human
The present LC-ESI-MS method was
employed to a crossover-design bio-
equivalencestudy
Human Ethics Committee of Institute of
Dermatonosis, Chinese
MedicalSciences
Twenty healthy male volunteers (mean
age 23.5 years, mean body mass index
21.8) were enrolled. Each participant
received a single dose (120 mg NTG) of
approved by the
Academyof
(Nanjing, China).
test (30 mg tablet, Wan Bang Medicine
Company, Nanjing, China) and refer-
ence products (120 mg tablet, Wan Bang
Medicine Company, Nanjing, China) in
a balance 2 9 2 Latin square design
experiment, separated by one week
washout period. The volunteers fasted
overnight prior to the administration of
NTG and 4 h post-dosing. Any medica-
tion, cigarette and drinks containing
caffeine or alcohol were not allowed at
least two weeks prior to and during the
periods of the test. Blood was obtained
before dosing and at 0.17, 0.33, 0.5, 0.75,
1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10 h after
administration.
Pharmacokinetics and
Statistical Analysis
Maximum plasma concentration (Cmax)
and time point of maximum plasma
concentration(Tmax)
directly from the measured data. The
half-life of drug elimination during
the terminal phase (t½), area under the
plasma concentration–time curve from 0
to the last measurable concentration
(AUC0–10h), area under the plasma con-
centration–time curve from 0 to infinity
(AUC0–?) and mean residence time
(MRT) were computed using WinNolin
5.0.1 (Pharsight corporation, Mountain
View, CA, USA). Statistical analysis of
bioequivalence parameters was carried
out using SPSS Software-Version 10.0. A
Wilcoxon test was established for Tmax.
For lnCmax, lnAUC0–10hand lnAUC0–?,
variance analysis was used to assess
period, personand
The two/one-side student’s t tests for
pharmacokineticparameters
lnAUC0–10h and lnAUC0–?) of NTG
were carried out. Bioequivalence was
assessed using a 90% confidence interval
(CI) for the ln-transformed bioequiva-
lence parameters, within an acceptable
range of 0.80–1.25 (www.fda.gov/cder/
guidance/index.htm, 2001).
wereobtained
producteffects.
(lnCmax,
Results
The LC–MS method described provided
good separation of NTG and Tinidazole
from
constituents. There was no interference
between the IS and NTG. Figure 1 shows
an MS chromatogram of blank human
plasma (a,b), blank human plasma con-
taining nateglinide and internal standard
(tinidazole) (c, d), and a plasma sample
from a volunteer administered a single
doseofnateglinidewithinternalstandard
(e, f). The retention times were about
4.1 min for NTG and 2.0 min for IS.
theotherendogenousplasma
Linearity and Limits
of Detection and Quantitation
Least
curves were constructed by plotting the
peak area ratios of analyte to IS versus
nominal concentration. The correlation
coefficient (r2) was typically better than
0.9993 over the concentration range of
0.05–16.0 lg mL-1.
between the nominal concentrations and
measured concentrations was generally
within 15% (the lowest concentration
within 20%) with the use of the least
squares regression. The lower limit of
quantification(LLOQ)was0.05 lgmL-1
for NTG in human plasma.
squaresregression calibration
The deviation
Extraction Recovery and
Matrix Effect
Results
absolute extraction recovery of nategli-
nide from human plasma are shown in
Table 1. No significant matrix effect was
observed. The negligible relative matrix
effect was also confirmed by the low
coefficient of variation between the
slopes of the five standard curves.
from determinationofthe
Precision and Accuracy
The data proved good precision and
accuracy of the method. As shown in
Table 1, the intra- and inter-day preci-
sions of the assay, as measured by the
coefficient of variation (RSD %), were
both lower than 10%. The accuracy
(RE) of intra-day and inter-day was also
lower than 10%. The results indicated
the method was reproducible.
Full Short CommunicationChromatographia 2010, 71, February (No. 3/4)
301

Page 4

Stability
Assessment of the stability of nategli-
nide in human plasma at room tem-
perature, -20 ?C for one month, and
after three freeze–thaw cycles was per-
formed. The results indicated that NTG
was stable in plasma under those con-
ditions
stability in the autosampler up for to
12 h was evaluated. The bias between
the concentrations determined from the
(Table 1). Processedsample
Fig. 1. Representative LC-ESI-MS positive SIM chromatograms obtained from blank human plasma at m/z 318.25 (a) and at m/z 247.90 (b), blank
human plasma spiked with Nateglinide (1 lg mL-1) and IS (25 lg mL-1) at m/z 318.25 (c) and m/z 247.90 (d), and a human plasma sample after
oral administration of 120 mg Nateglinide, monitored at m/z 318.25 (e) and m/z 247.90 (f)
302
Chromatographia 2010, 71, February (No. 3/4) Full Short Communication

Page 5

initial injection and the re-injections
was within 10%.
Bioequivalence Study
The main pharmacokinetic parameters
(Cmax, Tmax, AUC0–10h, AUC0–?, t½) of
NTG are shown in Table 2, and the
relative bioavailability of the test tablets
was 97.0 ± 8.5%. A Wilcoxon test was
established for Tmax and no statistical
difference was shown between the two
products (p > 0.05). The two/one-side
student’s t tests and variance analysis for
pharmacokinetic parameters
lnAUC0–10h and lnAUC0–?) of NTG
were calculated using SPSS 10.0. The
results indicated that there were no sig-
nificant differences in the periods and
products, and the two products of NTG
were bioequivalent in 90% confidential
limit.
(lnCmax,
Discussion and Conclusion
Our rapid, simple and sensitive LC–MS
method of NTG determination was fully
validated for specificity, sensitivity, pre-
cision and accuracy. The concentrations
of NTG used in the validation study
covered the need of our pharmacokinetic
study. The pharmacokinetic application
of this LC–MS method further proved
that it was sensitive and reproducible for
the pharmacokinetic study of NTG.
Several methods have been reported
for the quantification of NTG in bio-
logical fluids. These methods, which
utilized LC and UV detection with a
molecularly imprinted polymer [7], a
column-switching solid phase extraction
[8, 9], isotope radioactivity [4, 10], pre-
column fluorescence derivatization [15],
provided LLOQ of 10–50 ng mL-1. But
these methods were time-consuming,
involving multiple extraction steps, low
sensitivity and selectivity. Liquid chro-
matography–tandem mass spectrometry,
as a more definitive technique than UV
detection, may improve the sensitivity
and selectivity. In addition, tandem mass
spectrometry can provide more useful
sample information. But these methods
needed specialized
instruments which are of high instru-
and sophisticated
ment cost. Our method in the present
study provided a comparable LLOQ
value with a short run time of 4.1 min.
Moreover, sensitive and specific detec-
tion of NTG was achieved with less
sophisticated instruments.
In conclusion, a rapid, simple and
sensitiveLC–MSassaywasdevelopedand
validated, and had been successfully ap-
plied to a bioequivalence study for NTG.
Acknowledgements
This project was financially supported by
National ‘863’ Project (No. 2007AA02
Z171) and the National Natural Science
Foundation of China (No.30472060).
References
1. Valsamakis G, Kumar S (2000) Expert
Opin Pharmacother 1:1413–1421
2. Dunn CJ, Faulds D (2000) Drugs 60:607–
617
3. Fujitani S, Yada T (1994) Endocrinology
134:1395–1400
4. Weaver ML, Orwig BA, Rodriguez LC,
Graham ED, Chin JA, Shapiro MJ,
McLeod JF, Mangold JB (2001) Drug
Metab Dispos 29:415–421
5. Choudhury S, Hirschberg Y, Filipek R,
Lasseter K, McLeod JF (2000) J Clin
Pharmacol 40:634–640
6. Yan H, Yang G, Qiao F, Chen Y (2004) J
Pharm Biomed Anal 36:169–174
7. Yang G, Yin J, Li Z, Liu H, Cai L, Wang
D, Chen Y (2004) Chromatographia
59:705–709
8. Ono I, Matsuda K, Kanno S (1996) J
Chromatogr B Biomed Appl 678:384–387
Table 1. Validation of extraction recovery matrix effect intra-day inter-day precision and sta-
bility of nateglinide (n = 5)
Concentration
(lg mL-1)
Measured concentration
(mean ± SD)
RSD
(%)
Extraction recovery0.1
1.0
8.0
0.1
1.0
8.0
0.1
1.0
8.0
0.1
1.0
8.0
0.1
1.0
8.0
0.1
1.0
8.0
0.1
1.0
8.0
88.6 ± 6.5
96. 5 ± 0.7
103.4 ± 1.8
98.1 ± 6.3
94.8 ± 1.7
105.2 ± 1.0
0.1 ± 0.01
1.0 ± 0.05
7.9 ± 0.17
0.1 ± 0.01
1.0 ± 0.08
7.9 ± 0.37
0.11 ± 0.01
1.00 ± 0.05
7.82 ± 0.22
0.10 ± 0.01
1.04 ± 0.06
7.97 ± 0.30
0.10 ± 0.01
1.05 ± 0.04
7.75 ± 0.22
7.3
0.8
1.7
6.4
1.8
1.0
7.5
4.7
2.1
8.9
7.5
4.7
8.3
5.0
2.8
5.3
5.4
3.7
6.0
3.9
2.9
Matrix effect
Intra-day
Inter-day
Ambienta
Freezingb
Freeze–thawc
aMean ± standard deviation, average of concentration in 0 and 12 h
bMean ± standard deviation, average of concentration in 0 and 30 days
cMean ± standard deviation, average of concentration in 0, 1 and 2 cycles per week
Table 2. Pharmacokinetics parameters of nateglinide in 20 healthy male volunteers received a
single dose (120 mg nateglinide) of test and reference nateglinide tablets
ParametersReference (mean ± SD)Test (mean ± SD)
Cmax(lg mL-1)
Tmax(h)
t½(h)
MRT (h)
AUC0–10 h(lg h mL-1)
AUC0-?(lg h mL-1)
F (%)
4.17 ± 1.31
2.09 ± 1.06
1.93 ± 0.44
3.08 ± 0.65
13.67 ± 4.36
14.10 ± 4.58
4.37 ± 1.53
2.40 ± 0.97
1.92 ± 0.29
3.23 ± 0.65
13.21 ± 4.09
13.62 ± 4.31
97.4 ± 8.5
Full Short Communication Chromatographia 2010, 71, February (No. 3/4)
303